Improved Approaches and Algorithms for learning from Skewed Data Distribution using Cluster Disjuncts

Abstract

Data mining is the process of discovering hidden knowledge from newlinethe existing databases. In real-time applications, most often data sources are newlineof imbalanced nature. The traditional algorithms used for knowledge newlinediscovery are bottle necked due to wide range of data sources availability. newlineIn supervised classification, a new and burning challenge emerged for newlineresearcher community in data mining is Class Imbalance Learning (CIL). newlineThe problem of class imbalance learning is of great significance when newlinedealing with real-world datasets. The data imbalance problem is more newlineserious in the case of binary class, where the number of instances in one newlineclass predominantly outnumbers the number of instances in another class newline